ABSTRACT
The coronavirus disease (COVID-19) also known as SARS-Cov-2 has largely impacted the entire globe physically, economically, and psychologically. The detection of the virus in early stages is extremely crucial for faster recovery in patients and curbing its spread as its nature is highly contagious. Although several techniques are present today for the detection of coronavirus, they are laborious in nature, costly, require experts from medical science, and the accuracy is questionable in some of the traditional methods. This brings the need to search for a faster and reliable technique. Computer vision produced remarkable results in predicting the onset of various diseases, and the use of machine learning in healthcare has increased tremendously owing to the fast speed and high accuracy of results with minimal human intervention. Hence, this research paper aims to develop a computer vision-based artificial intelligence model that can predict the occurrence of coronavirus using electron microscopic images of the samples. In order to achieve the goal, YOLO v3 object detection algorithm using non-maxima suppression is used to classify whether a particular sample has coronavirus or not. It is proved that the proposed algorithm works faster than existing methodologies with considerably higher accuracy for detection of coronavirus. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.